To better understand the structural basis for folding, stability, metal binding, and reactivity of proteins and general and of Cu,Zn superoxide dismutase (SOD) specifically, we propose the renewal of our integrated program of SOD redesign, biochemical characterization, structural determination, and computational modeling. During the current funding period, we have determined crystallographic structures for wild-type human, bovine, and yeast SODs for designed mutant bovine and human enzymes, and for the human mitochondrial Mn superoxide dismutase, thereby providing an extensive database for the proposed analysis and design. Several mutant SODs have also been designed to test specific hypotheses regarding the basis for the enzyme's fold, stability, assembly metal binding, and electrostatic recognition of substrate. A recently designed mutant for medical applications created an SOD that binds to cell surface glycosaminoglycans, increasing SOD's serum half-life and ability to protect against vascular oxidative damage. The proposed work would continue these protein design, engineering and characterization studies, with an increased emphasis on x-ray structural determination and analysis SOD is an ideal model system, since our human bovine, and yeast SOD structures provide independent but related systems for testing structure function relationships, and the different crystal packing forms among these structures provided a basis for testing the interactions controlling crystal packing. Other advantages of studying SOD include the ability to obtain high-level expression in epsilon. coli and yeast, the select of SOD active molecules in SOD active molecules in SOD-minus epsilon. coli strains, and to test biological function by expressing our temperature-sensitive and other new SOD mutants in SOD-minus yeast strains and Drosophila clones. Furthermore, the interacting and selectively replaceable Cu+2 and Zn+2 sites in SOD allow controlled studies of metal site specificity and activity, the electrostatic guidance of the superoxide anion substrate provides a classic model for substrate recognition, and the beta-barrel structure provides a system for testing hypotheses regarding folding and evolution of this common protein fold by using component-based redesign. The unusually high subunit and dimer stability of SOD provides a basis for studying structural influences on protein assembly and stability. Genetic selection and the successes and failures of predictive modeling together will generate important new data for the improvement of current protein engineering methodology. Thus, the proposed research will increase our understanding of SOD structure and function at the atomic level, and we will both test and apply this new understanding by making specific structural changes to modify the folding, stability, assembly, substrate recognition, metal binding, and cell targeting in predetermined ways. In the broader sense, the proposed work aims to provide an improved fundamental basis for structure-based protein design and engineering that can be applied to many important proteins for scientific and medical purposes.